Current media delivery platforms, e.g., employed in adaptive streaming environments, are typically very rigid with regard to resources and configuration. Processing resources are allocated in advance for the processing of live or on-demand channels (i.e., ingestion, transcoding, encryption, segmentation and uploading), and once ingestion commences with respect to a media channel it may be difficult or impossible to alter the allocation/configuration of resources for that channel without risking undesirable disruption to downstream consumer viewing sessions. The current model for media delivery platforms does not allow the system to easily and efficiently adapt to variations in demand or underlying resource availability.